Customer Segmentation Using K-Means Clustering

# handle table-like data and matrices
import pandas as pd
import numpy as np

# visualisation
import seaborn as sns
import matplotlib.pyplot as plt
import missingno as msno
import plotly.express as px
import plotly.graph_objects as go
from plotly.subplots import make_subplots
import plotly.figure_factory as ff

# preprocessing
from sklearn.preprocessing import StandardScaler

# pca
from sklearn.decomposition import PCA

# clustering
from yellowbrick.cluster import KElbowVisualizer
from sklearn.cluster import KMeans, AgglomerativeClustering

# evaluations
from sklearn.metrics import confusion_matrix

# ignore warnings
import warnings
warnings.filterwarnings('ignore')

# to display the total number columns present in the dataset
pd.set_option('display.max_columns', None)

Import Data and handle missing values

Dataset Features

• ID: Customer's unique identifier
• Year_Birth: Customer's birth year
• Education: Customer's education level
• Marital_Status: Customer's marital status
• Income: Customer's yearly household income
• Kidhome: Number of children in customer's household
• Teenhome: Number of teenagers in customer's household
• Dt_Customer: Date of customer's enrollment with the company
• Recency: Number of days since customer's last purchase
• Complain: 1 if the customer complained in the last 2 years, 0 otherwise
• MntWines: Amount spent on wine in last 2 years
• MntFruits: Amount spent on fruits in last 2 years
• MntMeatProducts: Amount spent on meat in last 2 years
• MntFishProducts: Amount spent on fish in last 2 years
• MntSweetProducts: Amount spent on sweets in last 2 years
• MntGoldProds: Amount spent on gold in last 2 years
• NumDealsPurchases: Number of purchases made with a discount
• AcceptedCmp1: 1 if customer accepted the offer in the 1st campaign, 0 otherwise
• AcceptedCmp2: 1 if customer accepted the offer in the 2nd campaign, 0 otherwise
• AcceptedCmp3: 1 if customer accepted the offer in the 3rd campaign, 0 otherwise
• AcceptedCmp4: 1 if customer accepted the offer in the 4th campaign, 0 otherwise
• AcceptedCmp5: 1 if customer accepted the offer in the 5th campaign, 0 otherwise
• Response: 1 if customer accepted the offer in the last campaign, 0 otherwise
• NumWebPurchases: Number of purchases made through the company’s website
• NumCatalogPurchases: Number of purchases made using a catalogue
• NumStorePurchases: Number of purchases made directly in stores
• NumWebVisitsMonth: Number of visits to company’s website in the last month

df = pd.read_csv('marketing_campaign.csv', sep="\t")  

df.isnull().sum()

ID                      0
Year_Birth              0
Education               0
Marital_Status          0
Income                 24
Kidhome                 0
Teenhome                0
Dt_Customer             0
Recency                 0
MntWines                0
MntFruits               0
MntMeatProducts         0
MntFishProducts         0
MntSweetProducts        0
MntGoldProds            0
NumDealsPurchases       0
NumWebPurchases         0
NumCatalogPurchases     0
NumStorePurchases       0
NumWebVisitsMonth       0
AcceptedCmp3            0
AcceptedCmp4            0
AcceptedCmp5            0
AcceptedCmp1            0
AcceptedCmp2            0
Complain                0
Z_CostContact           0
Z_Revenue               0
Response                0
dtype: int64

msno.matrix(df)

<Axes: >

df = df.dropna()

df.isnull().sum()
     

ID                     0
Year_Birth             0
Education              0
Marital_Status         0
Income                 0
Kidhome                0
Teenhome               0
Dt_Customer            0
Recency                0
MntWines               0
MntFruits              0
MntMeatProducts        0
MntFishProducts        0
MntSweetProducts       0
MntGoldProds           0
NumDealsPurchases      0
NumWebPurchases        0
NumCatalogPurchases    0
NumStorePurchases      0
NumWebVisitsMonth      0
AcceptedCmp3           0
AcceptedCmp4           0
AcceptedCmp5           0
AcceptedCmp1           0
AcceptedCmp2           0
Complain               0
Z_CostContact          0
Z_Revenue              0
Response               0
dtype: int64

df.duplicated().sum()

0

Feature Engineering

df.info()

<class 'pandas.core.frame.DataFrame'>
Index: 2216 entries, 0 to 2239
Data columns (total 29 columns):
 #   Column               Non-Null Count  Dtype  
---  ------               --------------  -----  
 0   ID                   2216 non-null   int64  
 1   Year_Birth           2216 non-null   int64  
 2   Education            2216 non-null   object 
 3   Marital_Status       2216 non-null   object 
 4   Income               2216 non-null   float64
 5   Kidhome              2216 non-null   int64  
 6   Teenhome             2216 non-null   int64  
 7   Dt_Customer          2216 non-null   object 
 8   Recency              2216 non-null   int64  
 9   MntWines             2216 non-null   int64  
 10  MntFruits            2216 non-null   int64  
 11  MntMeatProducts      2216 non-null   int64  
 12  MntFishProducts      2216 non-null   int64  
 13  MntSweetProducts     2216 non-null   int64  
 14  MntGoldProds         2216 non-null   int64  
 15  NumDealsPurchases    2216 non-null   int64  
 16  NumWebPurchases      2216 non-null   int64  
 17  NumCatalogPurchases  2216 non-null   int64  
 18  NumStorePurchases    2216 non-null   int64  
 19  NumWebVisitsMonth    2216 non-null   int64  
 20  AcceptedCmp3         2216 non-null   int64  
 21  AcceptedCmp4         2216 non-null   int64  
 22  AcceptedCmp5         2216 non-null   int64  
 23  AcceptedCmp1         2216 non-null   int64  
 24  AcceptedCmp2         2216 non-null   int64  
 25  Complain             2216 non-null   int64  
 26  Z_CostContact        2216 non-null   int64  
 27  Z_Revenue            2216 non-null   int64  
 28  Response             2216 non-null   int64  
dtypes: float64(1), int64(25), object(3)
memory usage: 519.4+ KB

#Dt_Customer is not datetime
df['Dt_Customer'] = pd.to_datetime(df['Dt_Customer'], dayfirst=True)
df['Dt_Customer'] = pd.to_datetime(df['Dt_Customer'], format='mixed', dayfirst=True)

print("The newest customer's enrolment date in the records:", max(df['Dt_Customer']))
print("The oldest customer's enrolment date in the records:", min(df['Dt_Customer']))

The newest customer's enrolment date in the records: 2014-06-29 00:00:00
The oldest customer's enrolment date in the records: 2012-07-30 00:00:00

#Extract the age of the customer
df['Age'] = 2024 - df['Year_Birth']  

#Create a feature called Expenditure indicating the total amount spent by the customer in various categories over the span of two years.
df['Expenditure'] = df['MntWines'] + df['MntFruits'] + df['MntMeatProducts'] + df['MntFishProducts'] + df['MntSweetProducts'] + df['MntGoldProds']

#Create a feature called Stays_with indicating the living situation.
df['Stays_With'] = df['Marital_Status'].replace({'Married':'Partner', 'Together':'Partner', 'Absurd':'Alone', 'Widow':'Alone', 'YOLO':'Alone', 'Divorced':'Alone', 'Single':'Alone'})
  
#create a feature called offspring indicating the number of children the customer has.
df['Offspring'] = df['Kidhome'] + df['Teenhome'] 

#create a feature called family_size to find out the number of people living in the family.
df['Family_Size'] = df['Stays_With'].replace({'Alone': 1, 'Partner':2}) + df['Offspring']  

#Create a feature "Is_Parent" to indicate parenthood status
df['Is_Parent'] = np.where(df.Offspring > 0, 1, 0)  

#Segmentation of education level
df['Education'] = df['Education'].replace({'Basic':'Undergraduate', '2n Cycle':'Undergraduate', 'Graduation':'Graduate', 'Master':'Postgraduate', 'PhD':'Postgraduate'})
     

#Drop the features that are not important
drop = ['Marital_Status', 'Dt_Customer', 'Z_CostContact', 'Z_Revenue', 'Year_Birth', 'ID']
df = df.drop(drop, axis=1)
     

df.head()

	Education	Income	Kidhome	Teenhome	Recency	MntWines	MntFruits	MntMeatProducts	MntFishProducts	MntSweetProducts	MntGoldProds	NumDealsPurchases	NumWebPurchases	NumCatalogPurchases	NumStorePurchases	NumWebVisitsMonth	AcceptedCmp3	AcceptedCmp4	AcceptedCmp5	AcceptedCmp1	AcceptedCmp2	Complain	Response	Age	Expenditure	Stays_With	Offspring	Family_Size	Is_Parent
0	Graduate	58138.0	0	0	58	635	88	546	172	88	88	3	8	10	4	7	0	0	0	0	0	0	1	67	1617	Alone	0	1	0
1	Graduate	46344.0	1	1	38	11	1	6	2	1	6	2	1	1	2	5	0	0	0	0	0	0	0	70	27	Alone	2	3	1
2	Graduate	71613.0	0	0	26	426	49	127	111	21	42	1	8	2	10	4	0	0	0	0	0	0	0	59	776	Partner	0	2	0
3	Graduate	26646.0	1	0	26	11	4	20	10	3	5	2	2	0	4	6	0	0	0	0	0	0	0	40	53	Partner	1	3	1
4	Postgraduate	58293.0	1	0	94	173	43	118	46	27	15	5	5	3	6	5	0	0	0	0	0	0	0	43	422	Partner	1	3	1

Data Visualization

df.shape

(2216, 29)

df.describe()

	Income	Kidhome	Teenhome	Recency	MntWines	MntFruits	MntMeatProducts	MntFishProducts	MntSweetProducts	MntGoldProds	NumDealsPurchases	NumWebPurchases	NumCatalogPurchases	NumStorePurchases	NumWebVisitsMonth	AcceptedCmp3	AcceptedCmp4	AcceptedCmp5	AcceptedCmp1	AcceptedCmp2	Complain	Response	Age	Expenditure	Offspring	Family_Size	Is_Parent
count	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000	2216.000000
mean	52247.251354	0.441787	0.505415	49.012635	305.091606	26.356047	166.995939	37.637635	27.028881	43.965253	2.323556	4.085289	2.671029	5.800993	5.319043	0.073556	0.074007	0.073105	0.064079	0.013538	0.009477	0.150271	55.179603	607.075361	0.947202	2.592509	0.714350
std	25173.076661	0.536896	0.544181	28.948352	337.327920	39.793917	224.283273	54.752082	41.072046	51.815414	1.923716	2.740951	2.926734	3.250785	2.425359	0.261106	0.261842	0.260367	0.244950	0.115588	0.096907	0.357417	11.985554	602.900476	0.749062	0.905722	0.451825
min	1730.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	28.000000	5.000000	0.000000	1.000000	0.000000
25%	35303.000000	0.000000	0.000000	24.000000	24.000000	2.000000	16.000000	3.000000	1.000000	9.000000	1.000000	2.000000	0.000000	3.000000	3.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	47.000000	69.000000	0.000000	2.000000	0.000000
50%	51381.500000	0.000000	0.000000	49.000000	174.500000	8.000000	68.000000	12.000000	8.000000	24.500000	2.000000	4.000000	2.000000	5.000000	6.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	54.000000	396.500000	1.000000	3.000000	1.000000
75%	68522.000000	1.000000	1.000000	74.000000	505.000000	33.000000	232.250000	50.000000	33.000000	56.000000	3.000000	6.000000	4.000000	8.000000	7.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	65.000000	1048.000000	1.000000	3.000000	1.000000
max	666666.000000	2.000000	2.000000	99.000000	1493.000000	199.000000	1725.000000	259.000000	262.000000	321.000000	15.000000	27.000000	28.000000	13.000000	20.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000	131.000000	2525.000000	3.000000	5.000000	1.000000

df.describe(include=object).T

	count	unique	top	freq
Education	2216	3	Graduate	1116
Stays_With	2216	2	Partner	1430

#Plot of Expenditure , Income and Age of Customer
sns.pairplot(df , vars=['Expenditure','Income','Age'] , hue='Offspring', palette='husl')

<seaborn.axisgrid.PairGrid at 0x1f5f8307b90>

#Plot of Expenditure and Income of Customer
plt.figure(figsize=(13,8))
sns.scatterplot(x=df[df['Income']<600000]['Expenditure'], y=df[df['Income']<600000]['Income'], color='#f060a8')

<Axes: xlabel='Expenditure', ylabel='Income'>

#Plot of Expenditure and Age of Customer
plt.figure(figsize=(13,8))
sns.scatterplot(x=df['Expenditure'], y=df['Age'])

<Axes: xlabel='Expenditure', ylabel='Age'>

#Plot of Expenditure  and Education of Customer
plt.figure(figsize=(13,8))
sns.histplot(x=df['Expenditure'], hue=df['Education'])

<Axes: xlabel='Expenditure', ylabel='Count'>

#Plot of Education
df['Education'].value_counts().plot.pie(explode=[0.1,0,0], autopct='%1.1f%%', shadow=True, figsize=(8,8), colors=sns.color_palette('muted'))
     

<Axes: ylabel='count'>

Outlier Detection

#Plot of Age of Customer
plt.figure(figsize=(13,8))
sns.distplot(df.Age, color='#1943c2')

<Axes: xlabel='Age', ylabel='Density'>

#Plot of Income of Customer
plt.figure(figsize=(13,8))
sns.distplot(df.Income, color='#14db74')

<Axes: xlabel='Income', ylabel='Density'>

#Plot of Age of Customer
plt.figure(figsize=(13,8))
sns.distplot(df.Expenditure, color='#db1488')

<Axes: xlabel='Expenditure', ylabel='Density'>

#Detect outliers 
fig = make_subplots(rows=1, cols=3)

fig.add_trace(go.Box(y=df['Age'], notched=True, name='Age', marker_color = '#6699ff', 
                     boxmean=True, boxpoints='suspectedoutliers'), 1, 2)

fig.add_trace(go.Box(y=df['Income'], notched=True, name='Income', marker_color = '#ff0066', 
                     boxmean=True, boxpoints='suspectedoutliers'), 1, 1)

fig.add_trace(go.Box(y=df['Expenditure'], notched=True, name='Expenditure', marker_color = 'lightseagreen', 
                     boxmean=True, boxpoints='suspectedoutliers'), 1, 3)

fig.update_layout(title_text='Box Plots for Numerical Variables')

fig.show()
     

df.head(1)

	Education	Income	Kidhome	Teenhome	Recency	MntWines	MntFruits	MntMeatProducts	MntFishProducts	MntSweetProducts	MntGoldProds	NumDealsPurchases	NumWebPurchases	NumCatalogPurchases	NumStorePurchases	NumWebVisitsMonth	AcceptedCmp3	AcceptedCmp4	AcceptedCmp5	AcceptedCmp1	AcceptedCmp2	Complain	Response	Age	Expenditure	Stays_With	Offspring	Family_Size	Is_Parent
0	Graduate	58138.0	0	0	58	635	88	546	172	88	88	3	8	10	4	7	0	0	0	0	0	0	1	67	1617	Alone	0	1	0

numerical = ['Income', 'Recency', 'Age', 'Expenditure']

def detect_outliers(d):
  for i in d:
    Q3, Q1 = np.percentile(df[i], [75 ,25])
    IQR = Q3 - Q1

    ul = Q3+1.5*IQR
    ll = Q1-1.5*IQR

    outliers = df[i][(df[i] > ul) | (df[i] < ll)]
    print(f'*** {i} outlier points***', '\n', outliers, '\n')

detect_outliers(numerical)

*** Income outlier points*** 
 164     157243.0
617     162397.0
655     153924.0
687     160803.0
1300    157733.0
1653    157146.0
2132    156924.0
2233    666666.0
Name: Income, dtype: float64 

*** Recency outlier points*** 
 Series([], Name: Recency, dtype: int64) 

*** Age outlier points*** 
 192    124
239    131
339    125
Name: Age, dtype: int64 

*** Expenditure outlier points*** 
 1179    2525
1492    2524
1572    2525
Name: Expenditure, dtype: int64 

df = df[(df['Age']<100)]
df = df[(df['Income']<600000)]

df.shape

(2212, 29)

categorical = [var for var in df.columns if df[var].dtype=='O']
# check the number of different labels
for var in categorical:
    print(df[var].value_counts() / float(len(df)))
    print()
    print()

Education
Graduate         0.504069
Postgraduate     0.382007
Undergraduate    0.113924
Name: count, dtype: float64


Stays_With
Partner    0.64557
Alone      0.35443
Name: count, dtype: float64

categorical

['Education', 'Stays_With']

df['Stays_With'].unique()

array(['Alone', 'Partner'], dtype=object)

df['Education'].unique()

array(['Graduate', 'Postgraduate', 'Undergraduate'], dtype=object)

df['Education'] = df['Education'].map({'Undergraduate':0,'Graduate':1, 'Postgraduate':2})
     

df['Stays_With'] = df['Stays_With'].map({'Alone':0,'Partner':1})

df.dtypes

Education                int64
Income                 float64
Kidhome                  int64
Teenhome                 int64
Recency                  int64
MntWines                 int64
MntFruits                int64
MntMeatProducts          int64
MntFishProducts          int64
MntSweetProducts         int64
MntGoldProds             int64
NumDealsPurchases        int64
NumWebPurchases          int64
NumCatalogPurchases      int64
NumStorePurchases        int64
NumWebVisitsMonth        int64
AcceptedCmp3             int64
AcceptedCmp4             int64
AcceptedCmp5             int64
AcceptedCmp1             int64
AcceptedCmp2             int64
Complain                 int64
Response                 int64
Age                      int64
Expenditure              int64
Stays_With               int64
Offspring                int64
Family_Size              int64
Is_Parent                int32
dtype: object

df.head(3)

	Education	Income	Kidhome	Teenhome	Recency	MntWines	MntFruits	MntMeatProducts	MntFishProducts	MntSweetProducts	MntGoldProds	NumDealsPurchases	NumWebPurchases	NumCatalogPurchases	NumStorePurchases	NumWebVisitsMonth	AcceptedCmp3	AcceptedCmp4	AcceptedCmp5	AcceptedCmp1	AcceptedCmp2	Complain	Response	Age	Expenditure	Stays_With	Offspring	Family_Size	Is_Parent
0	1	58138.0	0	0	58	635	88	546	172	88	88	3	8	10	4	7	0	0	0	0	0	0	1	67	1617	0	0	1	0
1	1	46344.0	1	1	38	11	1	6	2	1	6	2	1	1	2	5	0	0	0	0	0	0	0	70	27	0	2	3	1
2	1	71613.0	0	0	26	426	49	127	111	21	42	1	8	2	10	4	0	0	0	0	0	0	0	59	776	1	0	2	0

corrmat = df.corr()

plt.figure(figsize=(20,20))  
sns.heatmap(corrmat, annot = True, cmap = 'crest', center = 0)

<Axes: >

Feature Scaling

df_old = df.copy()

# drop columns
drop_columns = ['AcceptedCmp3', 'AcceptedCmp4', 'AcceptedCmp5', 'AcceptedCmp1','AcceptedCmp2', 'Complain', 'Response']
df = df.drop(drop_columns , axis=1)

scaler = StandardScaler()
df = pd.DataFrame(scaler.fit_transform(df), columns = df.columns)

df.head(3)

	Education	Income	Kidhome	Teenhome	Recency	MntWines	MntFruits	MntMeatProducts	MntFishProducts	MntSweetProducts	MntGoldProds	NumDealsPurchases	NumWebPurchases	NumCatalogPurchases	NumStorePurchases	NumWebVisitsMonth	Age	Expenditure	Stays_With	Offspring	Family_Size	Is_Parent	Clusters
0	-0.411675	0.287105	-0.822754	-0.929699	0.310353	0.977660	1.552041	1.690293	2.453472	1.483713	0.852576	0.351030	1.426865	2.503607	-0.555814	0.692181	1.018352	1.676245	-1.349603	-1.264598	-1.758359	-1.581139	-0.310687
1	-0.411675	-0.260882	1.040021	0.908097	-0.380813	-0.872618	-0.637461	-0.718230	-0.651004	-0.634019	-0.733642	-0.168701	-1.126420	-0.571340	-1.171160	-0.132545	1.274785	-0.963297	-1.349603	1.404572	0.449070	0.632456	1.801732
2	-0.411675	0.913196	-0.822754	-0.929699	-0.795514	0.357935	0.570540	-0.178542	1.339513	-0.147184	-0.037254	-0.688432	1.426865	-0.229679	1.290224	-0.544908	0.334530	0.280110	0.740959	-1.264598	-0.654644	-1.581139	0.393452

Dimensionality Reduction

p = PCA(n_components=3)
p.fit(df)

PCA(n_components=3)

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

W = p.components_.T
W

array([[ 1.30778771e-02,  1.44277528e-01, -4.70771426e-01],
       [ 2.81248858e-01,  1.77878164e-01, -8.76658983e-02],
       [-2.45692674e-01,  5.76058529e-03,  2.69015093e-01],
       [-9.37194917e-02,  4.61594829e-01, -1.34887786e-01],
       [ 3.71636597e-03,  1.63414769e-02,  2.70181007e-02],
       [ 2.55668617e-01,  1.95331827e-01, -7.14503115e-02],
       [ 2.36504204e-01, -2.18575420e-03,  2.53321278e-01],
       [ 2.83905038e-01,  3.46528569e-05,  7.86274537e-02],
       [ 2.46940306e-01, -1.14624037e-02,  2.45744737e-01],
       [ 2.35683002e-01,  9.38301189e-03,  2.52354483e-01],
       [ 1.87969033e-01,  1.10934749e-01,  2.14824588e-01],
       [-7.54951077e-02,  3.39479300e-01,  1.88189814e-01],
       [ 1.68843410e-01,  2.83371310e-01,  6.95469695e-02],
       [ 2.76613344e-01,  9.47168407e-02,  3.11027304e-02],
       [ 2.41628747e-01,  1.92147752e-01,  2.15568134e-02],
       [-2.25384885e-01,  4.40103706e-02,  1.24000647e-01],
       [ 4.07595259e-02,  2.35408023e-01, -4.01417759e-01],
       [ 3.19061284e-01,  1.18345335e-01,  6.39589304e-02],
       [-2.73771544e-02,  1.15882801e-01,  2.90603792e-01],
       [-2.44084527e-01,  3.39331796e-01,  9.47820165e-02],
       [-2.16313970e-01,  3.41811001e-01,  2.31810203e-01],
       [-2.39720715e-01,  2.89828398e-01,  9.72583728e-02],
       [ 9.92652778e-02,  2.03177111e-01, -2.34086400e-01]])

pd.DataFrame(W, index=df.columns, columns=['W1','W2','W3'])

	W1	W2	W3
Education	0.013078	0.144278	-0.470771
Income	0.281249	0.177878	-0.087666
Kidhome	-0.245693	0.005761	0.269015
Teenhome	-0.093719	0.461595	-0.134888
Recency	0.003716	0.016341	0.027018
MntWines	0.255669	0.195332	-0.071450
MntFruits	0.236504	-0.002186	0.253321
MntMeatProducts	0.283905	0.000035	0.078627
MntFishProducts	0.246940	-0.011462	0.245745
MntSweetProducts	0.235683	0.009383	0.252354
MntGoldProds	0.187969	0.110935	0.214825
NumDealsPurchases	-0.075495	0.339479	0.188190
NumWebPurchases	0.168843	0.283371	0.069547
NumCatalogPurchases	0.276613	0.094717	0.031103
NumStorePurchases	0.241629	0.192148	0.021557
NumWebVisitsMonth	-0.225385	0.044010	0.124001
Age	0.040760	0.235408	-0.401418
Expenditure	0.319061	0.118345	0.063959
Stays_With	-0.027377	0.115883	0.290604
Offspring	-0.244085	0.339332	0.094782
Family_Size	-0.216314	0.341811	0.231810
Is_Parent	-0.239721	0.289828	0.097258
Clusters	0.099265	0.203177	-0.234086

pd.DataFrame(W, index=df.columns, columns=['W1','W2','W3'])

	W1	W2	W3
Education	0.013078	0.144278	-0.470771
Income	0.281249	0.177878	-0.087666
Kidhome	-0.245693	0.005761	0.269015
Teenhome	-0.093719	0.461595	-0.134888
Recency	0.003716	0.016341	0.027018
MntWines	0.255669	0.195332	-0.071450
MntFruits	0.236504	-0.002186	0.253321
MntMeatProducts	0.283905	0.000035	0.078627
MntFishProducts	0.246940	-0.011462	0.245745
MntSweetProducts	0.235683	0.009383	0.252354
MntGoldProds	0.187969	0.110935	0.214825
NumDealsPurchases	-0.075495	0.339479	0.188190
NumWebPurchases	0.168843	0.283371	0.069547
NumCatalogPurchases	0.276613	0.094717	0.031103
NumStorePurchases	0.241629	0.192148	0.021557
NumWebVisitsMonth	-0.225385	0.044010	0.124001
Age	0.040760	0.235408	-0.401418
Expenditure	0.319061	0.118345	0.063959
Stays_With	-0.027377	0.115883	0.290604
Offspring	-0.244085	0.339332	0.094782
Family_Size	-0.216314	0.341811	0.231810
Is_Parent	-0.239721	0.289828	0.097258
Clusters	0.099265	0.203177	-0.234086

p.explained_variance_

array([8.34736087, 3.0107941 , 1.4676905 ])

p.explained_variance_ratio_

array([0.36276466, 0.13084491, 0.06378378])

pd.DataFrame(p.explained_variance_ratio_, index=range(1,4), columns=['Explained Variability'])

	Explained Variability
1	0.362765
2	0.130845
3	0.063784

p.explained_variance_ratio_.cumsum()

array([0.99950708, 0.99991031, 0.99997959])

sns.barplot(x = list(range(1,4)), y = p.explained_variance_, palette = 'Paired')
plt.xlabel('i')
plt.ylabel('Lambda i')

Text(0, 0.5, 'Lambda i')

df_PCA = pd.DataFrame(p.transform(df), columns=(['col1', 'col2', 'col3']))
     

df_PCA.describe().T

	count	mean	std	min	25%	50%	75%	max
col1	2212.0	-1.220643e-16	2.889180	-5.972571	-2.593182	-0.747752	2.424818	7.444202
col2	2212.0	3.212219e-17	1.735164	-4.471689	-1.386241	-0.105978	1.308408	6.089454
col3	2212.0	-2.890997e-17	1.211485	-3.998819	-0.866740	-0.006152	0.828171	5.304697

x = df_PCA['col1']
y = df_PCA['col2']
z = df_PCA['col3']

fig = plt.figure(figsize=(13,8))
ax = fig.add_subplot(111, projection='3d')
ax.scatter(x,y,z, c='#e64363', marker='o')
ax.set_title('A 3D Projection of Data In the Reduced Dimension')
plt.show()
     

Train the model using K-means clustering

# Elbow Method to determine the number of clusters to be formed.
Elbow_M = KElbowVisualizer(KMeans(), k=10)
Elbow_M.fit(df_PCA)
Elbow_M.show()

<Axes: title={'center': 'Distortion Score Elbow for KMeans Clustering'}, xlabel='k', ylabel='distortion score'>

AC = AgglomerativeClustering(n_clusters=4)
# fit model and predict clusters
yhat_AC = AC.fit_predict(df_PCA)
df_PCA['Clusters'] = yhat_AC
#Adding the Clusters feature to the orignal dataframe.
df['Clusters'] = yhat_AC
df_old['Clusters'] = yhat_AC

fig = plt.figure(figsize=(13,8))
ax = plt.subplot(111, projection='3d', label='bla')
ax.scatter(x, y, z, s=40, c=df_PCA['Clusters'], marker='o', cmap='flare')
ax.set_title('Clusters')
plt.show()

plt.figure(figsize=(13,8))
pl = sns.countplot(x=df['Clusters'], palette= "viridis")
pl.set_title('Distribution Of The Clusters')
plt.show()

plt.figure(figsize=(13,8))
pl = sns.scatterplot(data=df_old, x=df_old['Expenditure'], y=df_old['Income'], hue=df_old['Clusters'], palette= "magma")
pl.set_title("Cluster's Profile Based on Income and Spending")
plt.legend()

<matplotlib.legend.Legend at 0x1f60835fb90>

Cluster Patterns

• group 0 is indicated in black and have average expenditure and average income
• group 1 is indicated in purple and have low expenditure and low income
• group 2 is indicated in orange and have high expenditure and average income
• group 3 is indicated in yellow and have high expenditure and high income

plt.figure(figsize=(13,8))
pl = sns.swarmplot(x=df_old['Clusters'], y=df_old['Expenditure'], color="#CBEDDD", alpha=0.7)
pl = sns.boxenplot(x=df_old['Clusters'], y=df_old['Expenditure'], palette= "rocket_r")
plt.show()

df_old['Total_Promos'] = df_old['AcceptedCmp1']+ df_old['AcceptedCmp2']+ df_old['AcceptedCmp3']+ df_old['AcceptedCmp4']+ df_old['AcceptedCmp5']

plt.figure(figsize=(13,8))
pl = sns.countplot(x=df_old['Total_Promos'], hue=df_old['Clusters'], palette= "viridis")
pl.set_title('Count Of Promotion Accepted')
pl.set_xlabel('Number Of Total Accepted Promotions')
plt.legend(loc='upper right')
plt.show()

plt.figure(figsize=(13,8))
pl=sns.boxenplot(y=df_old['NumDealsPurchases'],x=df_old['Clusters'],  palette= "viridis")
pl.set_title('Number of Deals Purchased')

Text(0.5, 1.0, 'Number of Deals Purchased')

plt.figure(figsize=(13,8))
sns.jointplot(x=df_old['Kidhome'], y=df_old['Expenditure'], hue=df_old['Clusters'], kind='kde', palette= "magma")

<seaborn.axisgrid.JointGrid at 0x1f609ed07d0>

<Figure size 1300x800 with 0 Axes>

plt.figure(figsize=(13,8))
sns.jointplot(x=df_old['Stays_With'], y=df_old['Expenditure'], hue=df_old['Clusters'], kind='kde', palette= "magma")

<seaborn.axisgrid.JointGrid at 0x1f60c14c790>

<Figure size 1300x800 with 0 Axes>

plt.figure(figsize=(13,8))
sns.jointplot(x=df_old['Education'], y=df_old['Expenditure'], hue=df_old['Clusters'], kind='kde', palette= "magma")

<seaborn.axisgrid.JointGrid at 0x1f60c6ccd10>

<Figure size 1300x800 with 0 Axes>

plt.figure(figsize=(13,8))
sns.jointplot(x=df_old['Is_Parent'], y=df_old['Expenditure'], hue=df_old['Clusters'], kind='kde', palette= "magma")

<seaborn.axisgrid.JointGrid at 0x1f608e9c590>

<Figure size 1300x800 with 0 Axes>

plt.figure(figsize=(13,8))
sns.jointplot(x=df_old['Family_Size'], y=df_old['Expenditure'], hue=df_old['Clusters'], kind='kde', palette= "magma")

<seaborn.axisgrid.JointGrid at 0x1f609ed7ed0>

<Figure size 1300x800 with 0 Axes>

plt.figure(figsize=(13,8))
sns.jointplot(x=df_old['Teenhome'], y=df_old['Expenditure'], hue=df_old['Clusters'], kind='kde', palette= "magma")

<seaborn.axisgrid.JointGrid at 0x1f609dfe350>

<Figure size 1300x800 with 0 Axes>

plt.figure(figsize=(13,8))
sns.jointplot(x=df_old['Age'], y=df_old['Expenditure'], hue=df_old['Clusters'], kind='kde', palette= "magma")

<seaborn.axisgrid.JointGrid at 0x1f609c586d0>

<Figure size 1300x800 with 0 Axes>

plt.figure(figsize=(13,8))
sns.jointplot(x=df_old['Offspring'], y=df_old['Expenditure'], hue=df_old['Clusters'], kind='kde', palette= "magma")

<seaborn.axisgrid.JointGrid at 0x1f5fd2fca50>

<Figure size 1300x800 with 0 Axes>

Inference:

• Cluster 0 has customers who are mostly parents with atleast 2 kids , with around five members in the family. Indicated by black

• Cluster 1 has customers who are mostly parents with one or no kid , with around of three members in the family. Indicated by purple.

• Cluster 2 has customers who are mostly parents with maximum of two kids , with around of four members in the family. Indicated by orange.

• Cluster 3 has customers not parents , with around of two members in the family. Indicated by yellow.